NEWS.md

In zajichek/cheese: Tools for Working with Data During Statistical Analysis

cheese 0.1.2 (in development)

• IN PROGRESS kableExtra::collapse_rows version 1.3.4 doesn't work properly causing issues for univariable_table HTML output. This release will require the latest version of that package once it is available on CRAN with the fix.

• IN PROGRESS Updating usage of dplyr functions to conform to most recent syntax best practices

• Updated functionality for cases when custom functions for "other" data types are used in univariate_table. Previously, these functions only picked up the summary values (i.e. it treated it as if it were a single summary statistic). Now, if the desired function produces a named vector as the result, the names (i.e. factor levels) will be attached and added to the result as is done for categorical data types.

cheese 0.1.1

Fixed a bug that caused the wrong ordering of columns in the result of univariate_table when there was more than nine (9) columns

cheese 0.1.0

Changes to existing functions

• absorb

• The print argument was renamed to trace

• descriptives

• The na.rm and useNA arguments are no longer required for additional functions

• Added na_string argument to for filling in NA result names
• Removed leading dots (e.g. .value) from result columns
• Added dedicated columns for the data type a function was evaluated for, and the names of a result

• Function, column, and result attributes have fun_, col_, and val_ prefixes, respectively

• divide

• Default behavior is now to split to a multi-level list depending on the number of split variables

• Control over the depth at which the list is split to
• Select split variables unquoted, quoted, or with select helpers as in dplyr::select()
• Control over whether unused split combinations are dropped with the drop argument

• Control over whether stratification variables are removed from the split frames with the remove argument

• stratiply

• User is in charge of binding back together (e.g. with fasten)

• stretch

• The keep argument is removed

• The keys argument is replaced with key and send argument replaced with value

• Extracting keys from the header is now done with grable().

• typly

• Removed the keep argument

• type_match

• Changed function name to some_type

• univariate_table

• Default access to 25th and 75th percentiles for string templates with "q1" and "q3", respectively

• Uses "median (q1, q3)" as the default string template for numeric variables
• Columns identified as other are not displayed by default
• Stratification variables are displayed as a hierarchy, and the sample size is always shown for the lowest group
• Row stratification variables are displayed consistently across rendering types to maintain headers

New functions

• depths finds the unique depth(s) of elements in a list that satisfy a predicate
• depths_string finds paths and locations of elements that satisfy a predicate
• fasten takes a divided data frame and binds it back together
• grable makes a knitr::kable with stacked headers
• muddle randomly permutes some or all of the columns of a data frame

cheese 0.0.3

Minor update to documentation

cheese 0.0.2

Minor update to documentation

cheese 0.0.1

Initial CRAN release

zajichek/cheese documentation built on July 26, 2023, 4:24 p.m.